Bindu Reddy

🚨 Agent Swarms - The Best Way To Turn Your AI Side-Hustle Into A $1B Business - use frontier LLMS to run multi-agent systems - combine Opus 4.7, GPT 5.5, Flash 3.5 and Grok 4.3 - create apps, marketing campaigns and automate sales - each agent specializes in different skills start your AI native business as a side-hustle and start making money

OpenAI can hit the ball out of the park…. By simply pricing 5.6 at half the price of 5.5 Imagine the number of serious developers who would move to the GPT class. Anthropic would be so over!

@thearslaniqbal they are correlated

@bindureddy Is the problem too much code, or too little understanding of what the code actually does?

We will continue to see layoffs from companies with large eng teams Engineers are producing 10-100x the code they used to.... This is causing a lot of instability in companies with large code bases and teams The status-quo is simply not sustainable

@PaulGugAI Well yeah, we will simply have a new normal That and a lot more people will vibe code a lot more things

I’m expecting there will be a transition however that transition will look different to the outcome. By that I mean continuing to see aggressive layoffs to recoup investment in AI. Following that- when the realisation sets in that AI didn’t automate everything it just moved the difficult parts around.. headcount will be increased again

TBH, Flash 3.5 is a fantastic chat model It’s very quick, pretty smart and a total joy to interact with Have been chatting with it all day

🚨 AGENT SWARMS - BUILD COMPLEX APPS AND AUTOMATIONS WITH ONE PROMPT Combine Gemini 3.1 Pro, Opus 4.7 and GPT 5.5 to create complex multi-agent systems Each agent excels at a particular task - coding, testing, mobile app, research and monitoring Master agent orchestrates worker agents - each agent uses the BEST AI for the task Cancel your SaaS subscriptions - Use AI to build custom apps

June is going to be an insane month - Gemini 3.5 Pro - GPT 5.6 and 5.6 Pro - Sonnet 4.7 - Mythos (maybe?) - Kimi or GLM next versions My current bet is on Google coming back, as long as they have reasonable prices

Google can still come back with 3.5 Pro, if they don’t increase prices A cheaper Opus class model will make a significant dent

The Flash Model That Rocks - DeepSeek Flash - It's dirt cheap. Almost 100x cheaper - works very well for small tasks - great for operating on large amounts of data/records

@somi_ai You switch to codex - and they all reset limits every 3 seconds anyways 😂😂

@bindureddy always-on agents on your own Claude keys means your rate limit is the real ceiling, not the supercompute. what happens when one hits the cap mid-task, does it queue or just drop it?

🚨 Deploy Hermes On The Abacus AI Super Computer You can now choose Claw or Hermes and deploy it on our cloud super computer - Bring you own Claude or Codex keys - Build on secure and scalable AI - Deploy in seconds Always on AI agents that you can simply message from Telegram or Whatsapp to do things

@ahmtaydn_9 yes

@bindureddy Can we use another api accept codex and claude? Is that allowed

Best Model For The Use Case Front-end coding - Opus 4.7 Back-end coding - GPT 5.5 xHigh Visual understanding- Flash 3.5 Cheap - DeepSeek Flash Video - Seedance 2.0 Image - GPT Image-2.0 Voice - Flash Live Writing - Gemini 3.1 Pro Real Time - Grok 4.3

@PJBolts70 Umm

@bindureddy at a lower wage, genius

Meta is a very weird company They will lay off 10000 people and in the next three months hire 10000 more 😲 Entire orgs are getting eliminated. Instead - keep the good people and just reassign them 🤷‍♀️

@mjzellinger lol wat / I literally have zero incentive to take any money from any frontier lab 😂😂

Now, as for you Bindu, the question I ask is if you have EVER engaged in any PAID or otherwise COMPENSATION marketing activity for ANY frontier lab (NOT LIMITED to ANTHROPIC, GOOGLE, OPENAI, GROK AI, OR THAT NEW LAB THAT JUST CAME OUT OF YC). I say this in such stringent tones because my lawyers are very strong and they have advised me not to take a soft posture when it comes to sophisticated actors like yourself, Bindu. That is all.

@vineelyalamarth Beg to differ.., I use all of them and would argue we are better

@bindureddy harness matters too, not just the model. cursor-2.5 is very good.

@Heeseon lol - that’s Kimi 2.6 🤷‍♀️

@bindureddy Composer 2 .5